vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Update ggml-cuda.cu

Browse source 
Co-authored-by: Jared Van Bortel <cebtenzzre@gmail.com>
This commit is contained in:
Andrew Godfrey 2023-11-15 18:58:05 -08:00 • committed by GitHub
parent e77094faab
commit bf46304cbc
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 0 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

1
ggml-cuda.cu
View file

@ -6250,7 +6250,6 @@ static int64_t get_row_rounding(ggml_type type) {
case GGML_TYPE_Q8_0: case GGML_TYPE_Q8_0:
return max_compute_capability >= CC_RDNA2 ? 128 : 64; return max_compute_capability >= CC_RDNA2 ? 128 : 64;
case GGML_TYPE_F16: case GGML_TYPE_F16:
return 1;
case GGML_TYPE_F32: case GGML_TYPE_F32:
return 1; return 1;
case GGML_TYPE_Q2_K: case GGML_TYPE_Q2_K: